Rules of Thumb for JTBD Practice

Key Principle

Collected heuristics convert ODI's theoretical framework into rapid practitioner decisions. Each rule encodes a lesson that, in the source material, required a full case study or analytical argument to derive. Practitioners who internalize these shortcuts avoid the most common and costly errors without re-deriving the logic each time.

Why This Matters

ODI is an 84-step process spanning qualitative research, quantitative survey design, statistical segmentation, and strategy selection. No practitioner can hold the full process in working memory during live decisions. Rules of thumb serve as compressed decision aids — they tell you when you are on track, when you have gone off course, and when a shortcut is safe versus dangerous.

These heuristics also function as diagnostic checks. When a project stalls or produces unexpected results, walking through the relevant rules of thumb often reveals the point of departure — a job scoped too narrowly, outcomes collected from the wrong informant, or a strategy mismatched to the segment's satisfaction state.

Good Examples

Job Definition Heuristics

State the job as verb + object + contextual clarifier, purely functional and solution-agnostic. (Chapter 4)
If the job definition references a product or technology, it is too narrow. "Boil water" lost to Keurig; "prepare a hot beverage for consumption" would have revealed the real competitive frame. (Chapter 4)
If the job yields too many outcomes to prioritize (200+), it may be too broad. 50-150 outcomes per job is the normal range; 200+ signals healthcare-level complexity or over-scoping. (Chapter 2)
Never ask "What job did you hire that product to do?" — this reveals only part of the job and encodes a product-centric bias. (Chapter 4)

Outcome Capture Heuristics

Collect outcomes from all three customer types: end user, product lifecycle support team, and purchase decision maker. Missing any one corrupts downstream analysis. (Chapter 4)
Interview the job executor, not the industry consultant. Arm & Hammer spent years getting input from nutritionists instead of dairy producers; technically sound products failed commercially. (Chapter 5)
A desired outcome statement must be measurable, solution-agnostic, stable over time, and controllable by the company. If it fails any of these tests, rewrite it. (Chapter 2)

Quantification Heuristics

Opportunity score = importance + max(importance - satisfaction, 0). Scores of 10+ indicate underserved outcomes worth targeting. (Chapters 3-4)
Never average importance/satisfaction scores across the full market without also segmenting. Bosch's circular saw market appeared mature on averages but hid a 30%+ underserved segment with 14 unmet outcomes. (Chapter 4)
Survey 180-3,000 customers for quantitative validation. (Chapter 7)

Strategy Selection Heuristics

Products that get the job done 20%+ better are "very likely to win in the marketplace." Below that threshold, switching costs often defeat the value proposition. (Chapter 2)
Never run a differentiated strategy against an overserved segment or a disruptive strategy against an underserved segment. Both are predictably fatal mismatches. (Chapter 3)
Sustaining strategies (<5% improvement) rarely justify new-entrant investment; they work only for incumbents defending existing position. (Chapter 3)

Organizational Heuristics

Separate "what to build" (small Innovation Center of Excellence) from "how to build" (rest of the organization). The bottleneck is target selection, not execution capability. (Chapter 7)
Build ODI capability around existing Six Sigma teams when available — they already have the process discipline and statistical fluency ODI demands. (Chapter 7)
ODI outcome data is a long-term asset. Because jobs and outcomes are stable, the research remains valid for years, not months. (Chapter 7)

Counterpoints

Rules of thumb can substitute for the full analytical process only on familiar territory. In novel markets with unfamiliar job structures, skipping the 84-step process and relying on heuristics alone reintroduces the "ideas-first" failure mode.
The 20% improvement threshold is an empirical guideline, not a universal law. Highly regulated or high-switching-cost markets may require larger deltas; low-friction digital markets may tolerate smaller ones.
"Never average without segmenting" does not mean segmentation is always necessary. If the full-market opportunity scores already reveal clear clusters of unmet needs (scores well above 10), segmentation confirms rather than reveals.
The "interview the executor, not the consultant" rule can overcorrect. Domain experts provide context for job mapping; they just should not be the source of desired outcome statements.

Key Quotes

"The mathematical probability of someone coming up with an idea that satisfactorily addresses all the customer's unmet needs without knowing what they are or whether or not they are satisfied is close to zero." — Ulwick, Chapter 1

"Making the core functional job the unit of analysis is the cornerstone of successful innovation." — Ulwick, Chapter 4

"Most companies are great at creating products — they just aren't that great at creating the right products." — Ulwick, Chapter 7

"Decisions are made using quantitative customer insights. Intuition is not acceptable." — Ulwick, Chapter 7

Rules of Thumb

Phase: Job Definition

State jobs as verb + object + contextual clarifier. No products, no technologies. (Ch. 4)
If your job definition could be disrupted by a platform shift, it is scoped too narrowly. (Ch. 4)
Expect 50-150 desired outcomes per job; 200+ signals over-scoping or extreme domain complexity. (Ch. 2)
All jobs decompose into up to eight universal steps: define, locate, prepare, confirm, execute, monitor, modify, conclude. (Ch. 4)

Phase: Outcome Capture 5. Interview all three customer types: executor, lifecycle support, purchase decision maker. (Ch. 4) 6. Source outcomes from the job executor, not adjacent experts or consultants. (Ch. 5) 7. Every outcome statement must pass four tests: measurable, solution-agnostic, stable, controllable. (Ch. 2) 8. Map the job (what the customer is trying to accomplish), not the journey (what they currently do). (Ch. 4)

Phase: Quantification & Segmentation 9. Opportunity score >= 10 means underserved. Target these first. (Ch. 3-4) 10. Always segment before concluding a market is mature. Hidden segments may contain 30%+ of the market. (Ch. 4) 11. Survey 180-3,000 customers for statistical validity. (Ch. 7)

Phase: Strategy & Execution 12. Target 20%+ improvement in job completion to trigger customer switching. (Ch. 2) 13. Match strategy to segment: differentiated for underserved, disruptive for overserved. Mismatches are predictably fatal. (Ch. 3) 14. Check whether existing products already address unmet outcomes before building new ones. Coloplast and Microsoft grew with messaging and repackaging, not new R&D. (Ch. 4-5) 15. Use outcome data as an internal persuasion tool — it replaces opinion-based debates with customer-validated evidence. (Ch. 5)